You are on page 1of 9

Revista de la Construccin

ISSN: 0717-7925
revistadelaconstruccion@uc.cl
Pontificia Universidad Catlica de Chile
Chile

Marn Uribe, Carlos Rodolfo; Thenoux Zeballos, Guillermo Alfonso


Validation of the Polygon-Of-Voids Tool for Asphalt Mixtures with RAP
Revista de la Construccin, vol. 13, nm. 1, abril, 2014, pp. 56-63
Pontificia Universidad Catlica de Chile
Santiago, Chile

Available in: http://www.redalyc.org/articulo.oa?id=127631777007

How to cite
Complete issue
More information about this article
Journal's homepage in redalyc.org

Scientific Information System


Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal
Non-profit academic project, developed under the open access initiative

[ 56

Validation of the Polygon-Of-Voids Tool for Asphalt Mixtures with RAP


Validacin de la herramienta de polgono de huecos para las mezclas asflticas con RAP
Carlos Rodolfo Marn Uribe (Autor Principal)

Guillermo Alfonso Thenoux Zeballos

Pontificia Universidad Catlica de Chile


Facultad de Ingeniera
Departamento de Ingeniera y Gestin de la Construccin
crmarin@uc.cl

Pontificia Universidad Catlica de Chile


Facultad de Ingeniera
Departamento de Ingeniera y Gestin de la Construccin
gthenoux@ing.puc.cl

Casilla 306, correo 22, Santiago de Chile.

Cdigo: 0214
Fecha de Recepcin: 01.01.2014.
Fecha de Aceptacin: 01.04.2014.

Abstract
Durability of an asphalt mixture is directly related to the optimum asphalt binder content, which guarantees coverage of the aggregate particles and
suitable volumetric properties to ensure a good performance in service, producing an asphalt mixture that is less susceptible to aging and moisture
damage. Previous research has shown that the polygon-of-voids tool, or polyvoids, is an analytical technique to determine the optimum asphalt binder
content of asphalt mixtures with virgin aggregates based on the specification limits of voids, allowing a saving in time and material. However, current
trends to develop new asphalt mixtures which incorporate reclaimed asphalt pavement (RAP) have created additional challenges given the nature of the
design of a mixture containing aggregates with residual asphalt that make the property measurement of the bulk specific gravity of the RAP aggregate
difficult. This paper aims to demonstrate the application of the polygon-of-voids tool for the design of recycled hot-mix asphalt. The results
show that the method of substitution, where the value of the actual specific gravity of the RAP aggregate
is assumed as the bulk specific gravity
of the RAP aggregate
, gives the optimum asphalt contents closest to those determined by the traditional Marshall method for the design of
asphalt mixtures.
Keywords: Reclaimed Asphalt Pavement (Rap); Polyvoids; Asphalt Mixes; Volumetric Properties.

INTRODUCTION
Generally, the design process of an asphalt mixture initially
consists of a volumetric design and subsequently of mechanical
or empirical tests to verify the design. The quality of the asphalt
mix can vary due to many factors such as variations in asphalt
binder content and particle size, which usually occur during
production in the plant, and variations in temperature and
compaction energy, which can happen during compaction in the
field, so that the compacted mixture on site may have different
volumetric parameters and mechanical properties from those
considered in the design (Garnica et al, 2005). It is desirable that
the mixture produced in the plant has uniform properties and
characteristics similar to those of the design of the laboratory
mix, which seeks to select the materials (aggregate, asphalt
binder, mineral filler, additives), the grading, the optimum
asphalt binder content, the mixing and compaction
temperatures, and the volumetric properties of the mixture, so
that the requirements chosen for a particular project are met.
The goal of the design process of hot-mix asphalt (HMA) is to
select a single optimum asphalt binder content that allows to
balance appropriately the properties looked for by the designer
and that ensures adequate service performance regarding
common failures such as fatigue, low temperature cracking and
permanent deformation. That is to say, to find the optimum
asphalt binder content which ensures a thickness of asphalt film
that delays aging of the mixture and subsequent cracking,
sufficient stiffness to meet the demands of traffic without
distorting or rutting, and convenient volumetric properties
which have historically been related to their performance and
durability. For any state of a compacted geo-material there are

2014, 13(1), 56 - 63

three properties of voids and its relation to the name of any


other geo-material is shown in Table 1.
Over the years, designers have established maximum and
minimum standards for these volumetric properties in order to
exclude those asphalt mixtures which have a poor performance.
For instance, low air voids may produce rutting and shove while
high air voids causes accelerated aging, high permeability,
brittleness, premature cracking, raveling and moisture damage
(McLeod, 1959; Asphalt Institute, 2007). The distribution of air
voids also affects the presence and movement of water in
asphalt mixtures. Water in asphalt mixture has harmful effects
on the pavement structure, which weakens the adhesion
between aggregates and asphalt and the cohesion of the
mixture itself, producing disintegration and subsequent failure
of the pavement structure. Voids in the mineral aggregate
quantify the area between the aggregate particles filled
with air and the effective asphalt binder content, control the
minimum asphalt binder content in the mixture and are
basically related to the asphalt covering of the aggregate
particles, the durability and stability (McLeod, 1956; Kandhal et
al, 1998; Attia et al, 2009).
is the measure used to ensure
proper asphalt film thickness on the aggregate particle for
acceptable durability of the mixture.

Table 1. Volumetric parameters of a geo-material. Source: Self Elaboration.

Properties of voids
Air voids
Voids in mineral aggregate
Voids filled with asphalt

Marn, C. Thenoux, G.

Generic name of a geomaterial


Total voids
Porosity
Degree of saturation

Revista de la Construccin
Journal of Construction

[ 57

PROBLEMS ASSOCIATED WITH THE DESIGN OF


HOT ASPHALT MIXTURES WITH RECLAIMED
ASPHALT PAVEMENT (RAP)
When it is necessary to determine the optimum asphalt binder
content and assess the volumetric properties of HMA with
contents of RAP, one of the most important properties that
must be determined is the effective specific gravity of the RAP
aggregate
. This parameter is critical to determine
accurately the volumetric parameters of the mixture, especially
the voids in the mineral aggregate
, a value that is one of
the key properties used for the design and the assurance of the
quality of the mixture, as mentioned above. Owing to the fact
that the bulk specific gravity of the RAP aggregate
cannot be measured directly, it is necessary to estimate it. If the
source of the RAP is known and the records of the original
building are available, the bulk specific gravity
value of
the virgin aggregate established in these records could be used
as value for the
. However, if these records (of the original
construction) are not available, the
value must be
estimated in accordance with one of the methodologies
historically proposed for this (see Table 2) (Mc Daniel, 2001;
NCHRP, 2001, Horan, 2003; Anderson, 2004, Al-Qadi et al.
2012).
The implication of this is that depending on the methodology
used to determine the
, the optimum asphalt binder
content might be different from the volumetric properties
found in the new HMA, a situation that could undermine a good
performance when in service. For example, in a mixture
containing 25% RAP, an error of 0.04 in the specific gravity can
affect the
calculated by approximately 0.5 percent. This
value can make a difference between a mix being accepted or
rejected in accordance with the quality requirements imposed,
or to the fatigue life being reduced significantly (Anderson,
2004).
Where:
= bulk specific gravity of RAP aggregate
= effective specific gravity of RAP aggregate
= theoretical maximum gravity of RAP aggregate (RICE)

Table 2. Methodologies for calculating the bulk specific gravity


aggregate. Source: Self Elaboration.

Methodology
Direct
Substitution

Back calculation

of the RAP

Description
Direct measurement of the
after being
recovered.
Estimate of the
through the value of the
. The
value is assumed as that of
the
.
Estimate of the
through the value of the
. The
value is assumed as that of
the
. Then, the asphalt absorption value
of the aggregates is assumed in order to
estimate the
.

The polygon of voids as an analytical tool for the design


of recycled asphalt mixes
The polygon of voids, or polyvoids, is an analytical tool used to
get the job formula or the optimum asphalt binder content for
any hot asphalt mixture based only on the specifications of
voids (Sanchez et al; 2011) and its rationale is supported by the
implementation of the gravimetric and volumetric relationships
of asphalt mixtures. Using this technique the volumetric
parameters of HMA, represented by the asphalt content
and the compacted density
, can be shown graphically
and analytically. That is to say, they determine the density of
the compacted asphalt
according to their volumetric
properties: air voids
, voids in the mineral aggregate
and voids filled with asphalt
and the constants of the
bulk specific gravity of the combined aggregate - virgin
aggregate and RAP aggregates , the effective
specific gravity of the combined aggregate
and the
specific gravity of the asphalt binder
. The formulation
proposed for construction of the polygon of voids is described
below, where the compacted density of the asphalt mixture is
defined in terms of air voids (see Table 3).
Figure 1 shows the construction of a polygon, using the
mathematical formulae described above. The intersection of
these curves, corresponding to the maximum and minimum
levels required by the specifications, gives rise to the vertices of
the polygon (Sanchez et al; 2011). There are at least ten
intersections that are converted into the key points for the
determination of the polygon of voids and they are shown in
Table 4. The definition of these points is given by the
intersections of the fundamental curves.

Table 3. Mathematical formulae for construction of the polygon of voids. Source: Self Elaboration.

Formula

Description
There is a specific situation when
. In this case, the equation represents
a limit because any combination of
and
may be found above this
curve called saturation curve and the equation becomes:
(2)
(
)

(1)

Similarly, the compacted density of an asphalt mixture is related to its voids


in the mineral aggregate of the following:
The proportion of voids filled with asphalt
is a degree of saturation,
that is, the relationship between the volume of voids filled with liquid and the
total volume of voids.
In the case that there is a 100% saturation, i.e.
, this equation
represents the saturation line and becomes:
(5)
(
)

(3)

2014, 13(1), 56 - 63

(4)

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction

[ 58
If one wanted to represent all the specifications of the voids for
a given asphalt mix within the plane
, there would
be a maximum area where all the specifications would be
complied with. This area is known as polygon of voids or
polyvoids. In accordance with this definition, any combination
of asphalt binder content and density inside the polygon should
simultaneously comply with all the void specifications (see
Figure 2). The job formula dealing with the content of optimum
asphalt binder is obtained by calculating the centroid of the
area of the polygon that the intersections of the curves of voids
generate (Sanchez et al, 2011).

because it gets the same results as a complete design with only


a fourth or a fifth of the number of samples, i.e. with
approximately three samples. As Sanchez has expressed in his
reports (2011), the savings due to the acceleration of the results
are not only for the benefit of the designers of asphalt mixtures
but also for that of researchers who state that applying this
technique can evaluate between four and five different
combinations of aggregates with the same resources and time
that would be used by applying the Marshall or Superpave
method in the conventional way.

Some previous work (Snchez-Leal, 2002; Snchez-Leal, 2004;


Sanchez et al, 2011) has demonstrated the application of this
tool in the analysis and design of HMA developed by both the
Marshall and Superpave methods with virgin materials.
However, due to the increase in the use of RAP in new asphalt
mixes, an interest has emerged to test whether the tool of
polyvoids allows to obtain the optimum asphalt binder content
and whether this result would be comparable to that obtained
with the traditional Marshall design proposed by the Asphalt
Institute (2007).

METHODOLOGY

Figure 1. Construction of the polygon of voids. Source: Self Elaboration.

Figure 3 shows the methodology proposed to carry out this


study, which presents two well-defined phases. The first one is
to implement the traditional Marshall design method for each
of the proposed mixtures and thus obtain the optimum asphalt
binder content. The second phase corresponds to the
application of the tool of polyvoids in accordance with the
characteristics of the materials and the design requirements of
the mixture in order to then also ascertain the optimum asphalt
binder content. This tool was implemented with the aid of a
spreadsheet programmed by the principal author.
Subsequently, the results were calibrated in accordance with
the data presented by Sanchez-Leal (2011) and compared with
those obtained by the traditional method.
Figure 2. Definition of the polygon of voids. Source: Self Elaboration.
Vertex
1
4.53
2.414
2
5.05
2.427
3
5.91
2.393
4
5.13
2.374
5
5.08
2.376
1
4.53
2.414
Centroid

Table 4. Key points of intersections of the fundamental curves of voids. Source:


Sanchez et al (2011).

Intersection N
1

(Va)max

(Va)min

(VMA)min

(Va)min

(VMA)max

(Va)max

(VMA)max

(VFA)min

(VMA)min

(VFA)max

(VMA)min

(VFA)max

(VMA)max

(VFA)min

(VMA)max

(VFA)max

(Va)min

10

(VFA)min

(Va)max

2.397

Figure 3. Methodology of the study. Source: Self Elaboration.

Intersection curve
(VMA)min

Advantages of the employment of the polygon of voids


The volumetric analysis in the Marshall as well as the Superpave
method for HMA design is performed by the production of
approximately 15 samples, i.e. three asphalt samples for each of
five different asphalt contents. The tool of polyvoids not only
makes a considerable saving in the design time of the asphalt
mixture possible but also in the number of samples used,
2014, 13(1), 56 - 63

5.1

Experimental Plan
During the first phase of the methodology, an experimental
plan was developed to apply the Marshall design to HMA with
different RAP contents (0%, 10%, 25%, 40%, 50% and 70%). For
the development of the entire design, various tasks were
carried out in order to characterize the materials used, define

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction

[ 59
the grading structure and obtain the optimum asphalt binder
content as is shown in Figure 4.

Figure 4. Flow chart of work for the design of asphalt mixtures with RAP. Source:
Self Elaboration.

Characterization of the virgin aggregate and RAP


The recycled material (RAP) has its origin in the milling of a layer
of asphalt in the street Colonel Souper in the Commune of
Santiago Centro (Santiago, Chile). Its characterization and that
of the virgin aggregate consists of three fundamental parts. The
first part corresponds to the separation of all aggregates,
including those of the RAP, according to size in the laboratory,
which helps to ensure a continuous distribution of particle sizes
when they are mixed in addition to facilitating the constitution
of an objective grading structure. These sizes are represented
by the sieves: " (12.7 mm), " (9.51 mm), 3/8" (9.5 mm), #4
(4.75 mm), #8 (2.38 mm), #30 (0,599mm), #50 (0,297mm), #100
(0,152 mm), # 200 (0,075 mm). In the second part the residual
asphalt content of RAP material is evaluated, and finally the
gravimetric properties of the aggregates are identified. Table 5
shows the laboratory tests carried out to define the three parts
of the characterization of the aggregates and Table 6 illustrates
the properties of the RAP aggregate.

Definition of the trial blend aggregate


Figure 5 shows the particle size chosen that corresponds to the
semi-dense type IV-A-12 established by the Manual of Roads of
Chile (2013) and recommended for surface layers.

Combination of the virgin aggregate and the RAP


The mixtures were designed while maintaining the same
grading regardless of the percentage of RAP added.
Subsequently, the values of the theoretical maximum gravity
in accordance with the standard AASTHO T209 were
obtained for all asphalt mixtures. Also, the
values were
estimated in accordance with the three methodologies
available.
The
combined
bulk
specific
gravity
measured of each blend with a different content of
RAP was determined by using the equation (6). Where is the
percentage of aggregate source and
is the aggregate bulk
specific gravity of source .

Table 5. Laboratory tests for the characterization of the aggregates. Source: Self
Elaboration.

Property

Standard test

Determination of the residual asphalt of the RAP


(centrifuge method)
Determination of the properties of the gravimetric
aggregates (virgin and retrieved from the RAP)
Specific Gravity and absorption of fine aggregates

AASHTO T 164
(Method A)

Specific Gravity and absorption of the coarse


aggregates
Specific weight of the solid

AASHTO T85

AASHTO T84

AASHTO T100

Table 6. RAP aggregate properties. Source: Self Elaboration.

Property
Residual asphalt content
Average of the theoretical maximum gravity

Standard test
5.5 %

Average of the bulk specific gravity


Direct method

2.643 gr/cm3

2.308 gr/cm3

Figure 5. Aggregate grading. Source: Self Elaboration.

Design of reclaimed asphalt mix using the Marshall


Method
The samples were manufactured by the Marshall method to 75
blows per side. This method was chosen given that it is still
current in Chile and the design of asphalt mixtures by the
Superpave method has not yet been implemented. An asphalt
of the type CA-24 was used, which is the one most commonly
used in the country. The design procedure was then followed
appropriately, producing three specimens for each of a total of
five asphalt contents, based on the initial content of asphalt
binder calculated using the procedure recommended by the
Asphalt Institute (2007), which was the starting point for the
complete design of the asphalt mixture (see Table 7).

(6)

Traditionally when a design of an asphalt mixture is carried out


by the Marshall method, the optimum amount of asphalt binder
2014, 13(1), 56 - 63

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction

[ 60
is obtained by determining the average of the three values that
represent the most important design properties: air voids
(generally 4%), maximum stability and maximum density.
However, it is current practice to calculate the optimum
amount of asphalt binder for the percentage of air voids
of
reference and then check the thresholds required for the other
properties of the mixture. This is the procedure that was
applied in this study and the results are shown in Table 8.

Table 7. Asphalt binder content according to the initial procedure of the asphalt
Institute MS-4. Source: Self Elaboration.

Content of RAP (%)

Initial binder content (%)

0
10
25
40
50
70

5.5
5.0
4.1
3.3
2.8
1.7

5%

5.4
5.2
5.6
5.4
5.7
6.1

5.1
4.9
5.3
5.0
6.0
5.8

Note: The percentages of optimum content of asphalt binder are established in


relation to the total weight of the mixture.

Table 9. Volumetric properties of control for obtaining the optimum asphalt binder
content. Source: Self Elaboration.

Volumetric
property

Symbol

Air voids

Target
value
4%
5%

Range

Observation

3% 5 %
4% 6%

This was also


calculated for
the
Chilean
conditions
for
nominal
maximum size
of
19
mm
according
to
Superpave.

Voids
in
mineral
aggregate

Minimum
13%

12% 14%

Voids filled
with asphalt

65% 80%

Generally, the HMA design requires a minimum and maximum


value for each one of the volumetric properties already defined.
Some researchers have recommended that the value of
is
limited to a maximum level, usually 1.5 - 2.0% above the
minimum value, in order to prevent a low resistance to rutting
(Christensen and Bonaquist 2006). Similarly, the
has
generally been established at 4%, ranging between 3 and 5%. In
Chile for example, the of design was established at 5% with a
range of between 4% and 6%. Table 9 shows the control values
for the volumetric properties used to establish the various job
formulae.

2014, 13(1), 56 - 63

Direct

Bulk Specific gravity of RAP


aggregate (gr/cm3)

0
10
25
40
50
70

It has been argued that if the


is incorrect, it will affect the
VMA calculated for the mixture, which could result in durability
issues. The magnitude of the VMA error will depend on the
error of the
(Kvasnak, 2010). Figure 6 illustrates the values
of
obtained from each methodology as well as the error
bars that indicate the 95% confidence interval. As can be seen, a
lower dispersion occurs in the values of bulk specific gravity
measured by the direct method while a greater dispersion was
obtained by the method of substitution, which may be due to
the sensitivity of the measurements for the determination of
.

2,700

Optimum content of asphalt binder for:


4%

Bulk Specific Gravity of the RAP aggregate

Figure 6. Bulk specific gravity of RAP aggregate. Source: Self Elaboration.

Table 8. Optimum content of asphalt binder (%) - Marshall method. Source: Self
Elaboration.

Amount of RAP in the


mix (%)

RESULTS AND DISCUSSION

2,650

Substitution
Back-calculation

2,600
2,550
2,500
2,450
2,400
2,350

Table 12. Optimum asphalt content for 4% of

Amount
of RAP in
the mix
(%)

Marshall

0
10
25
40
50
70

5.4
5.2
5.6
5.4
5.7
6.1

. Source: Self Elaboration.

Polyvoids
Direct

Substitution

5.1
4.8
4.9
4.4
4.9
4.5

5.1
5.0
5.4
5.3
6.0
6.2

Backcalculation
5.1
5.0
5.5
5.4
6.1
6.4

The results show that the


values obtained by the direct
method are higher while the other two methods give, on
average, similar results. This could be due to the extraction
process of the asphalt binder, which can change the properties
of the aggregate and may result in a change in the amount of
fine material, which can affect the specific gravity (NCHRP,
2001). In addition, the value of
found by the direct
method is greater than that obtained by the method of
substitution (where
is substituted by
), which is
inconsistent with what is mentioned by the NCHRP (2001),
which states the first property is always smaller than the second
one for a given aggregate.

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction

[ 61

Optimum asphalt binder content by Polyvoids

Figure 8. Percentage variation of optimum asphalt content measured through


three methodologies (for 4% ). Source: Self Elaboration.

7,0

As shown in Figure 7, the values closest to the optimum content


of asphalt binder obtained by the Marshall method, correspond
to those calculated by the substitution method of, followed by
those of the method of back-calculation and finally, very far off,
were the values of the direct method. The values obtained by
the method of back-calculation deviated from those produced
by the Marshall method from as much as 50% of RAP.

% Optimum asphalt content

Table 10 and 11 show the input data and the implementation of


the polygon of voids for the calculation of the optimum asphalt
binder content. This procedure was repeated for each of the
percentages of addition of RAP and the three methodologies of
calculating the value of
. The optimum contents of asphalt
binder for each of the asphalt mixtures with different contents
of RAP were obtained in the same way. Table 12 shows the
values that were established for 4% of .

Table 13 and Figure 9 show the values of optimum asphalt


binder content that were established for 5% of . It can be
seen that the values closest to the optimum content of asphalt
binder obtained by the Marshall method again correspond to
those calculated by the method of substitution, followed by
those of the back-calculation method and finally, very far off,
the values of the direct method. The optimum asphalt contents
determined by the method of back-calculation deviated from
those established by the Marshall method from 70% of RAP
onwards.

Figure 7. Polygon of voids construction for 4% of

. Source: Self Elaboration.

% Optimum asphalt content

7,0
6,5
6,0

6,0
5,5
5,0
4,5

4,0
0

10

25

40

50

70

% RAP
Marshall

Direct

Substitution

Figure 9. Polygon of voids construction for 5% of

% Optimal asphalt content

Given the dispersion of the values of


, one can calculate
the maximum and minimum asphalt binder content using the
polygon of voids and determine the respective ranges within
which this optimum asphalt binder can be found.

Figure 8 shows the average values of optimum asphalt content


and their respective limits according to the dispersion values of
calculated using the methodologies mentioned above. It
can be seen that as more RAP is added to the asphalt mixture,
the range of values of the optimum asphalt binder becomes
larger. This means that there is a higher dispersion of possible
values of optimum asphalt binder content when RAP aggregate
is added to the asphalt mixture. Therefore, the magnitude of
the
error will depend not only on error of the
measurement but also on the RAP content in the mixture.

6,5

Back-calculation

. Source: Self Elaboration.

6,5
6
5,5
5
4,5
4
0

10

20

30

40

50

60

70

80

% RAP
Marshall
Substitution

Table 13. Optimum asphalt content for 5% of

Amount of
RAP in the
mix (%)
0
10
25
40
50
70

Direct
Back-calculation

. Source: Self Elaboration.

Polyvoids
Marshall
5.1
4.9
5.3
5.0
6.0
5.8

Direct

Substitution

5.0
4.6
4.7
4.2
4.7
4.3

5.0
4.8
5.2
5.1
5.8
6.0

Backcalculation
5.0
4.8
5.3
5.3
6.0
6.2

CONCLUSION
The polygon of voids, or polyvoids, is an analytical tool used to
estimate the optimum content of asphalt binder making use
only of the volumetric properties of the asphalt mixture. This
tool has been tested successfully in asphalt mixes with virgin
aggregates according to the Marshall and Superpave methods.

5,5
5,0
4,5
4,0
0

10

20

Marshall
Substitution

2014, 13(1), 56 - 63

30
40
% RAP

50

60

70

Direct
Back-calculation

80

This work was intended to show the validity of the polygon of


voids as a predictor for the optimum content of asphalt binder
for hot mix asphalt with incorporation of varying amounts of
RAP, publicizing the difficulties that arose at the time of the
design. The results show that the above mentioned optimum
asphalt contents were similar to those estimated by means of a
typical Marshall design if the method of substitution is used for
the calculation of the bulk specific gravity of the aggregate of
the RAP
. This facilitates the preliminary estimate of the

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction

[ 62
asphalt content since the above method only requires the
calculation of the
by measurement of the
in the
laboratory, which is rather simple and quick to implement. In
the case of having the asphalt absorption property of the
aggregates, applying the method of back-calculation also gives
values close to those of the Marshall design, but they deviate as
more than 70% of RAP is incorporated in the new mix. One
advantage of this method is less dispersion in the calculation of
compared with the method of substitution, which allows
having certainty in determining the optimum asphalt binder
content. It is also important to establish that the property of the
air voids
was chosen as the criterion for the determination
of the optimum content of asphalt binder by the Marshall
method.
In summary, the tool of the polygon of voids appears promising
for estimating the optimum content of asphalt binder for
asphalt mixes containing reclaimed asphalt pavement.
However, for future research it is recommended to perform
more laboratory tests and confirm the findings of this research
using aggregates with different asphalt absorptions, RAP
sources, gradations and types of asphalt binder. Moreover, for
further validation of the polyvoids it is suggested to discuss in
detail about the dispersion of the values of
obtained by
the three methods in order to determine if there are changes in
the interpretation of the polygon of voids and volumetric
properties of the asphalt mixture.

REFERENCES
Al-Qadi, I; Aurangzeb, Q and Carpenter, (2012). Impact of high RAP
content on structural and performance properties of asphalt
mixtures. Research Report FHWA-ICT-12-002. Illinois Center
for Transportation, p.p: 3-4.
Anderson, M and Murphy, TR. (2004). Laboratory mix design using RAP:
Determining Aggregate Properties. www.asphaltmagazine.com;
Fall. Accessed: January, 2014.
Asphalt Institute. (2007). MS-4: The Asphalt Handbook (7th Edition).
ISBN13: 9781934154274.

Attia, MI; Abdelrahman, MA; Molakatalla, U; Salem, HM. (2009). Field


evaluation of asphalt film thickness as a design parameter in
Superpave mix design. International Journal of Pavement
Research Technology; 2(5):205-210.
Christensen, D.W; Jr; and Bonaquist, R.F. (2006). Volumetric
requirements for Superpave mix design. NCHRP rep. 567,
Transportation Research Board, Washington, D.C.

2014, 13(1), 56 - 63

Garnica, P., Flores, M, Gmez J.A and Delgado, H. (2005).


Caracterizacin geomecnica de mezclas asflticas.
Publicacin tcnica N 267, Instituto Mexicano del
Transporte (IMT), Sanfandila, Quertaro, Mxico, 2005.
Horan, R. (2003). Effective Specific Gravity for VMA Calculations.
www.asphaltmagazine.com; Fall 2003. Accessed: June 2014.
Kandal, P; Foo, K, and Mallick, R. (1998). Critical review of voids in
mineral aggregate requirements in Superpave. In
Transportation Research Record: Journal of the
Transportation Research Board, No. 1609, 21-27.
Kvasnak, A; West, R, Michael, J; Loria, L; Hajj, E and Tran, N. (2010). Bulk
Specific Gravity of Reclaimed Asphalt Pavement Aggregate.
Evaluating the Effect on Voids in Mineral Aggregate. In
Transportation Research Record: Journal of the
Transportation Research Board, No. 2180, 30-35.
Manual of Roads of Chile (2013). Direccin de Vialidad, Ministerio de
Obras Pblicas, Volumen No. 5: Especificaciones Tcnicas
Generales de Construccin, Edicin 2013, 265-268.
McDaniel, R. and Anderson, R.M. (2001). Recommended Use of
Reclaimed Asphalt Pavement in the Superpave Mix Design
Method: Technician's Manual, NCHRP Report No. 452,
Transportation Research Board, Washington, DC.
McLeod, N.W. (1956). Relationships between density, bitumen content
and voids properties of compacted bituminous paving
mixtures. Highway Research Board, Proc; 35th Anual
Meeting, vol 35, Washington, D.C, 327-404.
McLeod, N.W. (1959). Voids requirement for dense-graded bituminous
paving mixtures. Symp; Methods of Test for Design of
Bituminous Paving Mixtures, ASTM, West Conshohocken, PA,
81-112.
NCHRP (2001). Recommended Use of Reclaimed Asphalt Pavement in
the Superpave Mix Design Method: Guidelines. National
Cooperative Highway Research Program. Research Results
Digest, Number 253. Transportation Research Board,
National Research Council.
Snchez, F.J; Garnica, P; Larreal, M and Lpez, D.B. (2011). Polyvoids:
Analytical Tool for Superpave HMA Design. Journal of
Materials in Civil Engineering; vol 23, No. 8, 1129-1137.
Snchez-Leal, F.J. (2002). Nuevo enfoque para el diseo y control de
mezclas asflticas. Terceras Jornadas Internacionales del
Asfalto, Popayn, Colombia. Conference proceeding in
spanish.
Snchez-Leal, F.J.(2004). Entrenamiento especializado: RAMCODES
asistido por Origin. Application Manual, powered by Origin.
Software Shop Pro, Inc. Latinoamrica, Bogot, Colombia.

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction

[ 63
Table 10. Application of the tool of polyvoids to an asphalt mixture with 0% RAP. Source: Self Elaboration.

DIRECT METHOD

Vertex
1
2
3
4
5
1
Centroid

SUBSTITUTION METHOD

BACK-CALCULATION METHOD

Inputs
2.620

Inputs
2.620

Inputs
2.620

2.719

2.719

2.719

1.010

1.010

1.010

12

14

12

14

12

14

65

80

65

80

65

80

4.53
5.05
5.91
5.13
5.08
4.53
5.10

2.414
2.427
2.393
2.374
2.376
2.414
2.397

Vertex
1
2
3
4
5
1
Centroid

4.53
5.05
5.91
5.13
5.08
4.53
5.10

Vertex
1
2
3
4
5
1
Centroid

2.414
2.427
2.393
2.374
2.376
2.414
2.397
Gmb

Gmb

Gmb

2,420

2,420

2,400

2,400

2,380

2,380

2.414
2.427
2.393
2.374
2.376
2.414
2.397

2,440

2,440

2,440

4.53
5.05
5.91
5.13
5.08
4.53
5.10

2,420
2,400
2,380
2,360

2,360

2,360
4,0

4,5

5,0

5,5

6,0

6,5

4,0
4,0

4,5

5,0

5,5

6,0

Pb

4,5

5,0

5,5

6,0

6,5

Pb

6,5

Pb

Table 11. Application of the tool of polyvoids to an asphalt mixture with 25% RAP. Source: Self Elaboration.

Direct method

Substitution method

Inputs
2.625

Inputs
2.586

Inputs
2.580

2.697

2.697

2.697

1.020

Vertex
1
2
3
4
5
1
Centroid

Back-calculation method

1.020

1.020

12

14

12

14

12

14

65

80

65

80

65

80

4.24
4.76
5.64
4.84
4.82
4.24
4.85

2.412
2.425
2.391
2.372
2.372
2.412
2.395

Vertex
1
2
3
4
5
1
Centroid

4.81
5.33
6.21
5.42
5.34
4.81
5.44

Vertex
1
2
3
4
5
1
Centroid

2.390
2.403
2.369
2.350
2.353
2.390
2.373
Gmb

Gmb

Gmb

2,400

2,400

2,420
2,400
2,380

2,380

2,380

2,360

2,360
2,340

2,340

2,360
4,0

4,5

5,0

5,5

6,0

2.386
2.399
2.366
2.347
2.350
2.386
2.370

2,420

2,420

2,440

4.90
5.42
6.30
5.51
5.42
4.90
5.51

4,5

5,0

5,5

6,0

4,5

6,5

Pb

5,0

5,5

6,0

6,5

Pb

Pb

Notes:For asphalt mixtures from 25% to 70% RAP, an asphalt density of 1.020 gr/cm 3 was used in accordance with the recommendation made by Mc Daniels (2001). In this
study the residual asphalt density of RAP was not measure.

2014, 13(1), 56 - 63

Marn, C. Thenoux, G.

Revista de la Construccin
Journal of Construction